sql-glider 0.1.9__tar.gz → 0.1.10__tar.gz

{sql_glider-0.1.9 → sql_glider-0.1.10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sql-glider
-Version: 0.1.9
+Version: 0.1.10
 Summary: SQL Utility Toolkit for better understanding, use, and governance of your queries in a native environment.
 Project-URL: Homepage, https://github.com/rycowhi/sql-glider/
 Project-URL: Repository, https://github.com/rycowhi/sql-glider/

{sql_glider-0.1.9 → sql_glider-0.1.10}/src/sqlglider/_version.py

@@ -28,7 +28,7 @@ version_tuple: VERSION_TUPLE
 commit_id: COMMIT_ID
 __commit_id__: COMMIT_ID
 
-__version__ = version = '0.1.9'
-__version_tuple__ = version_tuple = (0, 1, 9)
+__version__ = version = '0.1.10'
+__version_tuple__ = version_tuple = (0, 1, 10)
 
 __commit_id__ = commit_id = None

{sql_glider-0.1.9 → sql_glider-0.1.10}/src/sqlglider/lineage/analyzer.py

@@ -751,7 +751,12 @@ class LineageAnalyzer:
 
         lineage_items = []
         # Get SQL for current expression only (not full multi-query SQL)
-        current_query_sql = self.expr.sql(dialect=self.dialect)
+        # For CACHE TABLE, pass just the SELECT since sqlglot.lineage doesn't
+        # natively understand CACHE statements
+        if isinstance(self.expr, exp.Cache) and self.expr.expression:
+            current_query_sql = self.expr.expression.sql(dialect=self.dialect)
+        else:
+            current_query_sql = self.expr.sql(dialect=self.dialect)
 
         for col in columns_to_analyze:
             try:

{sql_glider-0.1.9 → sql_glider-0.1.10}/tests/sqlglider/lineage/test_analyzer.py

@@ -2826,7 +2826,7 @@ class TestCacheTableStatements:
     """Tests for Spark SQL CACHE TABLE statement support."""
 
     def test_cache_table_as_select_column_lineage(self):
-        """CACHE TABLE t AS SELECT should produce lineage with t as target."""
+        """CACHE TABLE t AS SELECT should trace columns through to sources."""
         sql = """
         CACHE TABLE cached_customers AS
         SELECT customer_id, customer_name FROM customers
@@ -2835,10 +2835,27 @@ class TestCacheTableStatements:
         results = analyzer.analyze_queries(level=AnalysisLevel.COLUMN)
 
         assert len(results) == 1
-        output_names = [item.output_name for item in results[0].lineage_items]
-        assert any("cached_customers" in name for name in output_names)
-        assert any("customer_id" in name for name in output_names)
-        assert any("customer_name" in name for name in output_names)
+        items = {
+            item.output_name: item.source_name for item in results[0].lineage_items
+        }
+        assert items["cached_customers.customer_id"] == "customers.customer_id"
+        assert items["cached_customers.customer_name"] == "customers.customer_name"
+
+    def test_cache_lazy_table_column_lineage(self):
+        """CACHE LAZY TABLE should trace columns identically to CACHE TABLE."""
+        sql = """
+        CACHE LAZY TABLE cached_customers AS
+        SELECT customer_id, customer_name FROM customers
+        """
+        analyzer = LineageAnalyzer(sql, dialect="spark")
+        results = analyzer.analyze_queries(level=AnalysisLevel.COLUMN)
+
+        assert len(results) == 1
+        items = {
+            item.output_name: item.source_name for item in results[0].lineage_items
+        }
+        assert items["cached_customers.customer_id"] == "customers.customer_id"
+        assert items["cached_customers.customer_name"] == "customers.customer_name"
 
     def test_cache_table_as_select_table_extraction(self):
         """CACHE TABLE t AS SELECT should show cached_orders as OUTPUT table."""